Home

Context Navigation

source: palm/trunk/SOURCE/data_output_2d.f90 @ 3557

Last change on this file since 3557 was 3554, checked in by gronemeier, 6 years ago
renamed variable if to ivar; add variable description
Property svn:keywords set to `Id`
File size: 96.9 KB

Line
1	!> @file data_output_2d.f90
2	!------------------------------------------------------------------------------!
3	! This file is part of the PALM model system.
4	!
5	! PALM is free software: you can redistribute it and/or modify it under the
6	! terms of the GNU General Public License as published by the Free Software
7	! Foundation, either version 3 of the License, or (at your option) any later
8	! version.
9	!
10	! PALM is distributed in the hope that it will be useful, but WITHOUT ANY
11	! WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
12	! A PARTICULAR PURPOSE. See the GNU General Public License for more details.
13	!
14	! You should have received a copy of the GNU General Public License along with
15	! PALM. If not, see <http://www.gnu.org/licenses/>.
16	!
17	! Copyright 1997-2018 Leibniz Universitaet Hannover
18	!------------------------------------------------------------------------------!
19	!
20	! Current revisions:
21	! -----------------
22	!
23	!
24	! Former revisions:
25	! -----------------
26	! $Id: data_output_2d.f90 3554 2018-11-22 11:24:52Z eckhard $
27	! - add variable description
28	! - rename variable 'if' into 'ivar'
29	! - removed namelist LOCAL
30	! - removed variable rtext
31	!
32	! 3525 2018-11-14 16:06:14Z kanani
33	! Changes related to clean-up of biometeorology (dom_dwd_user)
34	!
35	! 3467 2018-10-30 19:05:21Z suehring
36	! Implementation of a new aerosol module salsa.
37	!
38	! 3448 2018-10-29 18:14:31Z kanani
39	! Add biometeorology
40	!
41	! 3421 2018-10-24 18:39:32Z gronemeier
42	! Renamed output variables
43	!
44	! 3419 2018-10-24 17:27:31Z gronemeier
45	! minor formatting (kanani)
46	! chem_data_output_2d subroutine added (basit)
47	!
48	! 3294 2018-10-01 02:37:10Z raasch
49	! changes concerning modularization of ocean option
50	!
51	! 3274 2018-09-24 15:42:55Z knoop
52	! Modularization of all bulk cloud physics code components
53	!
54	! 3241 2018-09-12 15:02:00Z raasch
55	! unused variables removed
56	!
57	! 3176 2018-07-26 17:12:48Z suehring
58	! Remove output of latent heat flux at urban-surfaces and set fill values
59	! instead
60	!
61	! 3052 2018-05-31 06:11:20Z raasch
62	! Do not open FORTRAN binary files in case of parallel netCDF I/O
63	!
64	! 3049 2018-05-29 13:52:36Z Giersch
65	! Error messages revised
66	!
67	! 3045 2018-05-28 07:55:41Z Giersch
68	! Error messages revised
69	!
70	! 3014 2018-05-09 08:42:38Z maronga
71	! Added nzb_do and nzt_do for some modules for 2d output
72	!
73	! 3004 2018-04-27 12:33:25Z Giersch
74	! precipitation_rate removed, case prr*_xy removed, to_be_resorted have to point
75	! to ql_vp_av and not to ql_vp, allocation checks implemented (averaged data
76	! will be assigned to fill values if no allocation happened so far)
77	!
78	! 2963 2018-04-12 14:47:44Z suehring
79	! Introduce index for vegetation/wall, pavement/green-wall and water/window
80	! surfaces, for clearer access of surface fraction, albedo, emissivity, etc. .
81	!
82	! 2817 2018-02-19 16:32:21Z knoop
83	! Preliminary gust module interface implemented
84	!
85	! 2805 2018-02-14 17:00:09Z suehring
86	! Consider also default-type surfaces for surface temperature output.
87	!
88	! 2797 2018-02-08 13:24:35Z suehring
89	! Enable output of ground-heat flux also at urban surfaces.
90	!
91	! 2743 2018-01-12 16:03:39Z suehring
92	! In case of natural- and urban-type surfaces output surfaces fluxes in W/m2.
93	!
94	! 2742 2018-01-12 14:59:47Z suehring
95	! Enable output of surface temperature
96	!
97	! 2735 2018-01-11 12:01:27Z suehring
98	! output of r_a moved from land-surface to consider also urban-type surfaces
99	!
100	! 2718 2018-01-02 08:49:38Z maronga
101	! Corrected "Former revisions" section
102	!
103	! 2696 2017-12-14 17:12:51Z kanani
104	! Change in file header (GPL part)
105	! Implementation of uv exposure model (FK)
106	! Implementation of turbulence_closure_mod (TG)
107	! Set fill values at topography grid points or e.g. non-natural-type surface
108	! in case of LSM output (MS)
109	!
110	! 2512 2017-10-04 08:26:59Z raasch
111	! upper bounds of cross section output changed from nx+1,ny+1 to nx,ny
112	! no output of ghost layer data
113	!
114	! 2292 2017-06-20 09:51:42Z schwenkel
115	! Implementation of new microphysic scheme: cloud_scheme = 'morrison'
116	! includes two more prognostic equations for cloud drop concentration (nc)
117	! and cloud water content (qc).
118	!
119	! 2277 2017-06-12 10:47:51Z kanani
120	! Removed unused variables do2d_xy_n, do2d_xz_n, do2d_yz_n
121	!
122	! 2233 2017-05-30 18:08:54Z suehring
123	!
124	! 2232 2017-05-30 17:47:52Z suehring
125	! Adjustments to new surface concept
126	!
127	!
128	! 2190 2017-03-21 12:16:43Z raasch
129	! bugfix for r2031: string rho replaced by rho_ocean
130	!
131	! 2031 2016-10-21 15:11:58Z knoop
132	! renamed variable rho to rho_ocean and rho_av to rho_ocean_av
133	!
134	! 2000 2016-08-20 18:09:15Z knoop
135	! Forced header and separation lines into 80 columns
136	!
137	! 1980 2016-07-29 15:51:57Z suehring
138	! Bugfix, in order to steer user-defined output, setting flag found explicitly
139	! to .F.
140	!
141	! 1976 2016-07-27 13:28:04Z maronga
142	! Output of radiation quantities is now done directly in the respective module
143	!
144	! 1972 2016-07-26 07:52:02Z maronga
145	! Output of land surface quantities is now done directly in the respective
146	! module
147	!
148	! 1960 2016-07-12 16:34:24Z suehring
149	! Scalar surface flux added
150	! Rename INTEGER variable s into s_ind, as s is already assigned to scalar
151	!
152	! 1849 2016-04-08 11:33:18Z hoffmann
153	! precipitation_amount, precipitation_rate, prr moved to arrays_3d
154	!
155	! 1822 2016-04-07 07:49:42Z hoffmann
156	! Output of bulk cloud physics simplified.
157	!
158	! 1788 2016-03-10 11:01:04Z maronga
159	! Added output of z0q
160	!
161	! 1783 2016-03-06 18:36:17Z raasch
162	! name change of netcdf routines and module + related changes
163	!
164	! 1745 2016-02-05 13:06:51Z gronemeier
165	! Bugfix: test if time axis limit exceeds moved to point after call of check_open
166	!
167	! 1703 2015-11-02 12:38:44Z raasch
168	! bugfix for output of single (*) xy-sections in case of parallel netcdf I/O
169	!
170	! 1701 2015-11-02 07:43:04Z maronga
171	! Bugfix in output of RRTGM data
172	!
173	! 1691 2015-10-26 16:17:44Z maronga
174	! Added output of Obukhov length (ol) and radiative heating rates for RRTMG.
175	! Formatting corrections.
176	!
177	! 1682 2015-10-07 23:56:08Z knoop
178	! Code annotations made doxygen readable
179	!
180	! 1585 2015-04-30 07:05:52Z maronga
181	! Added support for RRTMG
182	!
183	! 1555 2015-03-04 17:44:27Z maronga
184	! Added output of r_a and r_s
185	!
186	! 1551 2015-03-03 14:18:16Z maronga
187	! Added suppport for land surface model and radiation model output. In the course
188	! of this action, the limits for vertical loops have been changed (from nzb and
189	! nzt+1 to nzb_do and nzt_do, respectively in order to allow soil model output).
190	! Moreover, a new vertical grid zs was introduced.
191	!
192	! 1359 2014-04-11 17:15:14Z hoffmann
193	! New particle structure integrated.
194	!
195	! 1353 2014-04-08 15:21:23Z heinze
196	! REAL constants provided with KIND-attribute
197	!
198	! 1327 2014-03-21 11:00:16Z raasch
199	! parts concerning iso2d output removed,
200	! -netcdf output queries
201	!
202	! 1320 2014-03-20 08:40:49Z raasch
203	! ONLY-attribute added to USE-statements,
204	! kind-parameters added to all INTEGER and REAL declaration statements,
205	! kinds are defined in new module kinds,
206	! revision history before 2012 removed,
207	! comment fields (!:) to be used for variable explanations added to
208	! all variable declaration statements
209	!
210	! 1318 2014-03-17 13:35:16Z raasch
211	! barrier argument removed from cpu_log.
212	! module interfaces removed
213	!
214	! 1311 2014-03-14 12:13:39Z heinze
215	! bugfix: close #if defined( __netcdf )
216	!
217	! 1308 2014-03-13 14:58:42Z fricke
218	! +local_2d_sections, local_2d_sections_l, ns
219	! Check, if the limit of the time dimension is exceeded for parallel output
220	! To increase the performance for parallel output, the following is done:
221	! - Update of time axis is only done by PE0
222	! - Cross sections are first stored on a local array and are written
223	! collectively to the output file by all PEs.
224	!
225	! 1115 2013-03-26 18:16:16Z hoffmann
226	! ql is calculated by calc_liquid_water_content
227	!
228	! 1076 2012-12-05 08:30:18Z hoffmann
229	! Bugfix in output of ql
230	!
231	! 1065 2012-11-22 17:42:36Z hoffmann
232	! Bugfix: Output of cross sections of ql
233	!
234	! 1053 2012-11-13 17:11:03Z hoffmann
235	! +qr, nr, qc and cross sections
236	!
237	! 1036 2012-10-22 13:43:42Z raasch
238	! code put under GPL (PALM 3.9)
239	!
240	! 1031 2012-10-19 14:35:30Z raasch
241	! netCDF4 without parallel file support implemented
242	!
243	! 1007 2012-09-19 14:30:36Z franke
244	! Bugfix: missing calculation of ql_vp added
245	!
246	! 978 2012-08-09 08:28:32Z fricke
247	! +z0h
248	!
249	! Revision 1.1 1997/08/11 06:24:09 raasch
250	! Initial revision
251	!
252	!
253	! Description:
254	! ------------
255	!> Data output of cross-sections in netCDF format or binary format
256	!> to be later converted to NetCDF by helper routine combine_plot_fields.
257	!> Attention: The position of the sectional planes is still not always computed
258	!> --------- correctly. (zu is used always)!
259	!------------------------------------------------------------------------------!
260	SUBROUTINE data_output_2d( mode, av )
261
262
263	USE arrays_3d, &
264	ONLY: dzw, e, heatflux_output_conversion, nc, nr, p, pt, &
265	precipitation_amount, prr, q, qc, ql, ql_c, ql_v, qr, s, tend, &
266	u, v, vpt, w, zu, zw, waterflux_output_conversion, hyrho, d_exner
267
268	USE averaging
269
270	USE basic_constants_and_equations_mod, &
271	ONLY: c_p, lv_d_cp, l_v
272
273	USE biometeorology_mod, &
274	ONLY: bio_data_output_2d
275
276	USE bulk_cloud_model_mod, &
277	ONLY: bulk_cloud_model, bcm_data_output_2d
278
279	USE chemistry_model_mod, &
280	ONLY: chem_data_output_2d
281
282	USE control_parameters, &
283	ONLY: air_chemistry, biometeorology, data_output_2d_on_each_pe, &
284	data_output_xy, data_output_xz, data_output_yz, do2d, &
285	do2d_xy_last_time, do2d_xy_time_count, &
286	do2d_xz_last_time, do2d_xz_time_count, &
287	do2d_yz_last_time, do2d_yz_time_count, &
288	ibc_uv_b, io_blocks, io_group, land_surface, message_string, &
289	ntdim_2d_xy, ntdim_2d_xz, ntdim_2d_yz, &
290	ocean_mode, psolver, section, simulated_time, &
291	time_since_reference_point, uv_exposure
292
293	USE cpulog, &
294	ONLY: cpu_log, log_point
295
296	USE gust_mod, &
297	ONLY: gust_data_output_2d, gust_module_enabled
298
299	USE indices, &
300	ONLY: nbgp, nx, nxl, nxlg, nxr, nxrg, ny, nyn, nyng, nys, nysg, &
301	nzb, nzt, wall_flags_0
302
303	USE kinds
304
305	USE land_surface_model_mod, &
306	ONLY: lsm_data_output_2d, zs
307
308	#if defined( __netcdf )
309	USE NETCDF
310	#endif
311
312	USE netcdf_interface, &
313	ONLY: fill_value, id_set_xy, id_set_xz, id_set_yz, id_var_do2d, &
314	id_var_time_xy, id_var_time_xz, id_var_time_yz, nc_stat, &
315	netcdf_data_format, netcdf_handle_error
316
317	USE ocean_mod, &
318	ONLY: ocean_data_output_2d
319
320	USE particle_attributes, &
321	ONLY: grid_particles, number_of_particles, particle_advection_start, &
322	particles, prt_count
323
324	USE pegrid
325
326	USE radiation_model_mod, &
327	ONLY: radiation, radiation_data_output_2d
328
329	USE salsa_mod, &
330	ONLY: salsa, salsa_data_output_2d
331
332	USE surface_mod, &
333	ONLY: ind_pav_green, ind_veg_wall, ind_wat_win, surf_def_h, &
334	surf_lsm_h, surf_usm_h
335
336	USE turbulence_closure_mod, &
337	ONLY: tcm_data_output_2d
338
339	USE uv_exposure_model_mod, &
340	ONLY: uvem_data_output_2d
341
342
343	IMPLICIT NONE
344
345	CHARACTER (LEN=2) :: do2d_mode !< output mode of variable ('xy', 'xz', 'yz')
346	CHARACTER (LEN=2) :: mode !< mode with which the routine is called ('xy', 'xz', 'yz')
347	CHARACTER (LEN=4) :: grid !< string defining the vertical grid
348
349	INTEGER(iwp) :: av !< flag for (non-)average output
350	INTEGER(iwp) :: ngp !< number of grid points of an output slice
351	INTEGER(iwp) :: file_id !< id of output files
352	INTEGER(iwp) :: flag_nr !< number of masking flag
353	INTEGER(iwp) :: i !< loop index
354	INTEGER(iwp) :: iis !< vertical index of a xy slice in array 'local_2d_sections'
355	INTEGER(iwp) :: is !< slice index
356	INTEGER(iwp) :: ivar !< variable index
357	INTEGER(iwp) :: j !< loop index
358	INTEGER(iwp) :: k !< loop index
359	INTEGER(iwp) :: l !< loop index
360	INTEGER(iwp) :: layer_xy !< vertical index of a xy slice in array 'local_pf'
361	INTEGER(iwp) :: m !< loop index
362	INTEGER(iwp) :: n !< loop index
363	INTEGER(iwp) :: nis !< number of vertical slices to be written via parallel NetCDF output
364	INTEGER(iwp) :: ns !< number of output slices
365	INTEGER(iwp) :: nzb_do !< lower limit of the data field (usually nzb)
366	INTEGER(iwp) :: nzt_do !< upper limit of the data field (usually nzt+1)
367	INTEGER(iwp) :: s_ind !< index of slice types (xy=1, xz=2, yz=3)
368	INTEGER(iwp) :: sender !< PE id of sending PE
369	INTEGER(iwp) :: ind(4) !< index limits (lower/upper bounds) of array 'local_2d'
370
371	LOGICAL :: found !< true if output variable was found
372	LOGICAL :: resorted !< true if variable is resorted
373	LOGICAL :: two_d !< true if variable is only two dimensional
374
375	REAL(wp) :: mean_r !< mean particle radius
376	REAL(wp) :: s_r2 !< sum( particle-radius**2 )
377	REAL(wp) :: s_r3 !< sum( particle-radius**3 )
378
379	REAL(wp), DIMENSION(:), ALLOCATABLE :: level_z !< z levels for output array
380	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: local_2d !< local 2-dimensional array containing output values
381	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: local_2d_l !< local 2-dimensional array containing output values
382
383	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_pf !< output array
384	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_2d_sections !< local array containing values at all slices
385	REAL(wp), DIMENSION(:,:,:), ALLOCATABLE :: local_2d_sections_l !< local array containing values at all slices
386
387	#if defined( __parallel )
388	REAL(wp), DIMENSION(:,:), ALLOCATABLE :: total_2d !< same as local_2d
389	#endif
390	REAL(wp), DIMENSION(:,:,:), POINTER :: to_be_resorted !< points to array which shall be output
391
392	!
393	!-- Immediate return, if no output is requested (no respective sections
394	!-- found in parameter data_output)
395	IF ( mode == 'xy' .AND. .NOT. data_output_xy(av) ) RETURN
396	IF ( mode == 'xz' .AND. .NOT. data_output_xz(av) ) RETURN
397	IF ( mode == 'yz' .AND. .NOT. data_output_yz(av) ) RETURN
398
399	CALL cpu_log (log_point(3),'data_output_2d','start')
400
401	two_d = .FALSE. ! local variable to distinguish between output of pure 2D
402	! arrays and cross-sections of 3D arrays.
403
404	!
405	!-- Depending on the orientation of the cross-section, the respective output
406	!-- files have to be opened.
407	SELECT CASE ( mode )
408
409	CASE ( 'xy' )
410	s_ind = 1
411	ALLOCATE( level_z(nzb:nzt+1), local_2d(nxl:nxr,nys:nyn) )
412
413	IF ( netcdf_data_format > 4 ) THEN
414	ns = 1
415	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
416	ns = ns + 1
417	ENDDO
418	ns = ns - 1
419	ALLOCATE( local_2d_sections(nxl:nxr,nys:nyn,1:ns) )
420	local_2d_sections = 0.0_wp
421	ENDIF
422
423	!
424	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
425	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
426	CALL check_open( 101+av*10 )
427	ENDIF
428	IF ( data_output_2d_on_each_pe .AND. netcdf_data_format < 5 ) THEN
429	CALL check_open( 21 )
430	ELSE
431	IF ( myid == 0 ) THEN
432	#if defined( __parallel )
433	ALLOCATE( total_2d(0:nx,0:ny) )
434	#endif
435	ENDIF
436	ENDIF
437
438	CASE ( 'xz' )
439	s_ind = 2
440	ALLOCATE( local_2d(nxl:nxr,nzb:nzt+1) )
441
442	IF ( netcdf_data_format > 4 ) THEN
443	ns = 1
444	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
445	ns = ns + 1
446	ENDDO
447	ns = ns - 1
448	ALLOCATE( local_2d_sections(nxl:nxr,1:ns,nzb:nzt+1) )
449	ALLOCATE( local_2d_sections_l(nxl:nxr,1:ns,nzb:nzt+1) )
450	local_2d_sections = 0.0_wp; local_2d_sections_l = 0.0_wp
451	ENDIF
452
453	!
454	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
455	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
456	CALL check_open( 102+av*10 )
457	ENDIF
458
459	IF ( data_output_2d_on_each_pe .AND. netcdf_data_format < 5 ) THEN
460	CALL check_open( 22 )
461	ELSE
462	IF ( myid == 0 ) THEN
463	#if defined( __parallel )
464	ALLOCATE( total_2d(0:nx,nzb:nzt+1) )
465	#endif
466	ENDIF
467	ENDIF
468
469	CASE ( 'yz' )
470	s_ind = 3
471	ALLOCATE( local_2d(nys:nyn,nzb:nzt+1) )
472
473	IF ( netcdf_data_format > 4 ) THEN
474	ns = 1
475	DO WHILE ( section(ns,s_ind) /= -9999 .AND. ns <= 100 )
476	ns = ns + 1
477	ENDDO
478	ns = ns - 1
479	ALLOCATE( local_2d_sections(1:ns,nys:nyn,nzb:nzt+1) )
480	ALLOCATE( local_2d_sections_l(1:ns,nys:nyn,nzb:nzt+1) )
481	local_2d_sections = 0.0_wp; local_2d_sections_l = 0.0_wp
482	ENDIF
483
484	!
485	!-- Parallel netCDF4/HDF5 output is done on all PEs, all other on PE0 only
486	IF ( myid == 0 .OR. netcdf_data_format > 4 ) THEN
487	CALL check_open( 103+av*10 )
488	ENDIF
489
490	IF ( data_output_2d_on_each_pe .AND. netcdf_data_format < 5 ) THEN
491	CALL check_open( 23 )
492	ELSE
493	IF ( myid == 0 ) THEN
494	#if defined( __parallel )
495	ALLOCATE( total_2d(0:ny,nzb:nzt+1) )
496	#endif
497	ENDIF
498	ENDIF
499
500	CASE DEFAULT
501	message_string = 'unknown cross-section: ' // TRIM( mode )
502	CALL message( 'data_output_2d', 'PA0180', 1, 2, 0, 6, 0 )
503
504	END SELECT
505
506	!
507	!-- For parallel netcdf output the time axis must be limited. Return, if this
508	!-- limit is exceeded. This could be the case, if the simulated time exceeds
509	!-- the given end time by the length of the given output interval.
510	IF ( netcdf_data_format > 4 ) THEN
511	IF ( mode == 'xy' .AND. do2d_xy_time_count(av) + 1 > &
512	ntdim_2d_xy(av) ) THEN
513	WRITE ( message_string, * ) 'Output of xy cross-sections is not ', &
514	'given at t=', simulated_time, '&because the', &
515	' maximum number of output time levels is exceeded.'
516	CALL message( 'data_output_2d', 'PA0384', 0, 1, 0, 6, 0 )
517	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
518	RETURN
519	ENDIF
520	IF ( mode == 'xz' .AND. do2d_xz_time_count(av) + 1 > &
521	ntdim_2d_xz(av) ) THEN
522	WRITE ( message_string, * ) 'Output of xz cross-sections is not ', &
523	'given at t=', simulated_time, '&because the', &
524	' maximum number of output time levels is exceeded.'
525	CALL message( 'data_output_2d', 'PA0385', 0, 1, 0, 6, 0 )
526	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
527	RETURN
528	ENDIF
529	IF ( mode == 'yz' .AND. do2d_yz_time_count(av) + 1 > &
530	ntdim_2d_yz(av) ) THEN
531	WRITE ( message_string, * ) 'Output of yz cross-sections is not ', &
532	'given at t=', simulated_time, '&because the', &
533	' maximum number of output time levels is exceeded.'
534	CALL message( 'data_output_2d', 'PA0386', 0, 1, 0, 6, 0 )
535	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
536	RETURN
537	ENDIF
538	ENDIF
539
540	!
541	!-- Allocate a temporary array for resorting (kji -> ijk).
542	ALLOCATE( local_pf(nxl:nxr,nys:nyn,nzb:nzt+1) )
543	local_pf = 0.0
544
545	!
546	!-- Loop of all variables to be written.
547	!-- Output dimensions chosen
548	ivar = 1
549	l = MAX( 2, LEN_TRIM( do2d(av,ivar) ) )
550	do2d_mode = do2d(av,ivar)(l-1:l)
551
552	DO WHILE ( do2d(av,ivar)(1:1) /= ' ' )
553
554	IF ( do2d_mode == mode ) THEN
555	!
556	!-- Set flag to steer output of radiation, land-surface, or user-defined
557	!-- quantities
558	found = .FALSE.
559
560	nzb_do = nzb
561	nzt_do = nzt+1
562	!
563	!-- Before each output, set array local_pf to fill value
564	local_pf = fill_value
565	!
566	!-- Set masking flag for topography for not resorted arrays
567	flag_nr = 0
568
569	!
570	!-- Store the array chosen on the temporary array.
571	resorted = .FALSE.
572	SELECT CASE ( TRIM( do2d(av,ivar) ) )
573	CASE ( 'e_xy', 'e_xz', 'e_yz' )
574	IF ( av == 0 ) THEN
575	to_be_resorted => e
576	ELSE
577	IF ( .NOT. ALLOCATED( e_av ) ) THEN
578	ALLOCATE( e_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
579	e_av = REAL( fill_value, KIND = wp )
580	ENDIF
581	to_be_resorted => e_av
582	ENDIF
583	IF ( mode == 'xy' ) level_z = zu
584
585	CASE ( 'thetal_xy', 'thetal_xz', 'thetal_yz' )
586	IF ( av == 0 ) THEN
587	to_be_resorted => pt
588	ELSE
589	IF ( .NOT. ALLOCATED( lpt_av ) ) THEN
590	ALLOCATE( lpt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
591	lpt_av = REAL( fill_value, KIND = wp )
592	ENDIF
593	to_be_resorted => lpt_av
594	ENDIF
595	IF ( mode == 'xy' ) level_z = zu
596
597	CASE ( 'lwp*_xy' ) ! 2d-array
598	IF ( av == 0 ) THEN
599	DO i = nxl, nxr
600	DO j = nys, nyn
601	local_pf(i,j,nzb+1) = SUM( ql(nzb:nzt,j,i) * &
602	dzw(1:nzt+1) )
603	ENDDO
604	ENDDO
605	ELSE
606	IF ( .NOT. ALLOCATED( lwp_av ) ) THEN
607	ALLOCATE( lwp_av(nysg:nyng,nxlg:nxrg) )
608	lwp_av = REAL( fill_value, KIND = wp )
609	ENDIF
610	DO i = nxl, nxr
611	DO j = nys, nyn
612	local_pf(i,j,nzb+1) = lwp_av(j,i)
613	ENDDO
614	ENDDO
615	ENDIF
616	resorted = .TRUE.
617	two_d = .TRUE.
618	level_z(nzb+1) = zu(nzb+1)
619
620	CASE ( 'ghf*_xy' ) ! 2d-array
621	IF ( av == 0 ) THEN
622	DO m = 1, surf_lsm_h%ns
623	i = surf_lsm_h%i(m)
624	j = surf_lsm_h%j(m)
625	local_pf(i,j,nzb+1) = surf_lsm_h%ghf(m)
626	ENDDO
627	DO m = 1, surf_usm_h%ns
628	i = surf_usm_h%i(m)
629	j = surf_usm_h%j(m)
630	local_pf(i,j,nzb+1) = surf_usm_h%frac(ind_veg_wall,m) * &
631	surf_usm_h%wghf_eb(m) + &
632	surf_usm_h%frac(ind_pav_green,m) * &
633	surf_usm_h%wghf_eb_green(m) + &
634	surf_usm_h%frac(ind_wat_win,m) * &
635	surf_usm_h%wghf_eb_window(m)
636	ENDDO
637	ELSE
638	IF ( .NOT. ALLOCATED( ghf_av ) ) THEN
639	ALLOCATE( ghf_av(nysg:nyng,nxlg:nxrg) )
640	ghf_av = REAL( fill_value, KIND = wp )
641	ENDIF
642	DO i = nxl, nxr
643	DO j = nys, nyn
644	local_pf(i,j,nzb+1) = ghf_av(j,i)
645	ENDDO
646	ENDDO
647	ENDIF
648
649	resorted = .TRUE.
650	two_d = .TRUE.
651	level_z(nzb+1) = zu(nzb+1)
652
653	CASE ( 'ol*_xy' ) ! 2d-array
654	IF ( av == 0 ) THEN
655	DO m = 1, surf_def_h(0)%ns
656	i = surf_def_h(0)%i(m)
657	j = surf_def_h(0)%j(m)
658	local_pf(i,j,nzb+1) = surf_def_h(0)%ol(m)
659	ENDDO
660	DO m = 1, surf_lsm_h%ns
661	i = surf_lsm_h%i(m)
662	j = surf_lsm_h%j(m)
663	local_pf(i,j,nzb+1) = surf_lsm_h%ol(m)
664	ENDDO
665	DO m = 1, surf_usm_h%ns
666	i = surf_usm_h%i(m)
667	j = surf_usm_h%j(m)
668	local_pf(i,j,nzb+1) = surf_usm_h%ol(m)
669	ENDDO
670	ELSE
671	IF ( .NOT. ALLOCATED( ol_av ) ) THEN
672	ALLOCATE( ol_av(nysg:nyng,nxlg:nxrg) )
673	ol_av = REAL( fill_value, KIND = wp )
674	ENDIF
675	DO i = nxl, nxr
676	DO j = nys, nyn
677	local_pf(i,j,nzb+1) = ol_av(j,i)
678	ENDDO
679	ENDDO
680	ENDIF
681	resorted = .TRUE.
682	two_d = .TRUE.
683	level_z(nzb+1) = zu(nzb+1)
684
685	CASE ( 'p_xy', 'p_xz', 'p_yz' )
686	IF ( av == 0 ) THEN
687	IF ( psolver /= 'sor' ) CALL exchange_horiz( p, nbgp )
688	to_be_resorted => p
689	ELSE
690	IF ( .NOT. ALLOCATED( p_av ) ) THEN
691	ALLOCATE( p_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
692	p_av = REAL( fill_value, KIND = wp )
693	ENDIF
694	IF ( psolver /= 'sor' ) CALL exchange_horiz( p_av, nbgp )
695	to_be_resorted => p_av
696	ENDIF
697	IF ( mode == 'xy' ) level_z = zu
698
699	CASE ( 'pc_xy', 'pc_xz', 'pc_yz' ) ! particle concentration
700	IF ( av == 0 ) THEN
701	IF ( simulated_time >= particle_advection_start ) THEN
702	tend = prt_count
703	! CALL exchange_horiz( tend, nbgp )
704	ELSE
705	tend = 0.0_wp
706	ENDIF
707	DO i = nxl, nxr
708	DO j = nys, nyn
709	DO k = nzb, nzt+1
710	local_pf(i,j,k) = tend(k,j,i)
711	ENDDO
712	ENDDO
713	ENDDO
714	resorted = .TRUE.
715	ELSE
716	IF ( .NOT. ALLOCATED( pc_av ) ) THEN
717	ALLOCATE( pc_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
718	pc_av = REAL( fill_value, KIND = wp )
719	ENDIF
720	! CALL exchange_horiz( pc_av, nbgp )
721	to_be_resorted => pc_av
722	ENDIF
723
724	CASE ( 'pr_xy', 'pr_xz', 'pr_yz' ) ! mean particle radius (effective radius)
725	IF ( av == 0 ) THEN
726	IF ( simulated_time >= particle_advection_start ) THEN
727	DO i = nxl, nxr
728	DO j = nys, nyn
729	DO k = nzb, nzt+1
730	number_of_particles = prt_count(k,j,i)
731	IF (number_of_particles <= 0) CYCLE
732	particles => grid_particles(k,j,i)%particles(1:number_of_particles)
733	s_r2 = 0.0_wp
734	s_r3 = 0.0_wp
735	DO n = 1, number_of_particles
736	IF ( particles(n)%particle_mask ) THEN
737	s_r2 = s_r2 + particles(n)%radius*2 &
738	particles(n)%weight_factor
739	s_r3 = s_r3 + particles(n)%radius*3 &
740	particles(n)%weight_factor
741	ENDIF
742	ENDDO
743	IF ( s_r2 > 0.0_wp ) THEN
744	mean_r = s_r3 / s_r2
745	ELSE
746	mean_r = 0.0_wp
747	ENDIF
748	tend(k,j,i) = mean_r
749	ENDDO
750	ENDDO
751	ENDDO
752	! CALL exchange_horiz( tend, nbgp )
753	ELSE
754	tend = 0.0_wp
755	ENDIF
756	DO i = nxl, nxr
757	DO j = nys, nyn
758	DO k = nzb, nzt+1
759	local_pf(i,j,k) = tend(k,j,i)
760	ENDDO
761	ENDDO
762	ENDDO
763	resorted = .TRUE.
764	ELSE
765	IF ( .NOT. ALLOCATED( pr_av ) ) THEN
766	ALLOCATE( pr_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
767	pr_av = REAL( fill_value, KIND = wp )
768	ENDIF
769	! CALL exchange_horiz( pr_av, nbgp )
770	to_be_resorted => pr_av
771	ENDIF
772
773	CASE ( 'theta_xy', 'theta_xz', 'theta_yz' )
774	IF ( av == 0 ) THEN
775	IF ( .NOT. bulk_cloud_model ) THEN
776	to_be_resorted => pt
777	ELSE
778	DO i = nxl, nxr
779	DO j = nys, nyn
780	DO k = nzb, nzt+1
781	local_pf(i,j,k) = pt(k,j,i) + lv_d_cp * &
782	d_exner(k) * &
783	ql(k,j,i)
784	ENDDO
785	ENDDO
786	ENDDO
787	resorted = .TRUE.
788	ENDIF
789	ELSE
790	IF ( .NOT. ALLOCATED( pt_av ) ) THEN
791	ALLOCATE( pt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
792	pt_av = REAL( fill_value, KIND = wp )
793	ENDIF
794	to_be_resorted => pt_av
795	ENDIF
796	IF ( mode == 'xy' ) level_z = zu
797
798	CASE ( 'q_xy', 'q_xz', 'q_yz' )
799	IF ( av == 0 ) THEN
800	to_be_resorted => q
801	ELSE
802	IF ( .NOT. ALLOCATED( q_av ) ) THEN
803	ALLOCATE( q_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
804	q_av = REAL( fill_value, KIND = wp )
805	ENDIF
806	to_be_resorted => q_av
807	ENDIF
808	IF ( mode == 'xy' ) level_z = zu
809
810	CASE ( 'ql_xy', 'ql_xz', 'ql_yz' )
811	IF ( av == 0 ) THEN
812	to_be_resorted => ql
813	ELSE
814	IF ( .NOT. ALLOCATED( ql_av ) ) THEN
815	ALLOCATE( ql_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
816	ql_av = REAL( fill_value, KIND = wp )
817	ENDIF
818	to_be_resorted => ql_av
819	ENDIF
820	IF ( mode == 'xy' ) level_z = zu
821
822	CASE ( 'ql_c_xy', 'ql_c_xz', 'ql_c_yz' )
823	IF ( av == 0 ) THEN
824	to_be_resorted => ql_c
825	ELSE
826	IF ( .NOT. ALLOCATED( ql_c_av ) ) THEN
827	ALLOCATE( ql_c_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
828	ql_c_av = REAL( fill_value, KIND = wp )
829	ENDIF
830	to_be_resorted => ql_c_av
831	ENDIF
832	IF ( mode == 'xy' ) level_z = zu
833
834	CASE ( 'ql_v_xy', 'ql_v_xz', 'ql_v_yz' )
835	IF ( av == 0 ) THEN
836	to_be_resorted => ql_v
837	ELSE
838	IF ( .NOT. ALLOCATED( ql_v_av ) ) THEN
839	ALLOCATE( ql_v_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
840	ql_v_av = REAL( fill_value, KIND = wp )
841	ENDIF
842	to_be_resorted => ql_v_av
843	ENDIF
844	IF ( mode == 'xy' ) level_z = zu
845
846	CASE ( 'ql_vp_xy', 'ql_vp_xz', 'ql_vp_yz' )
847	IF ( av == 0 ) THEN
848	IF ( simulated_time >= particle_advection_start ) THEN
849	DO i = nxl, nxr
850	DO j = nys, nyn
851	DO k = nzb, nzt+1
852	number_of_particles = prt_count(k,j,i)
853	IF (number_of_particles <= 0) CYCLE
854	particles => grid_particles(k,j,i)%particles(1:number_of_particles)
855	DO n = 1, number_of_particles
856	IF ( particles(n)%particle_mask ) THEN
857	tend(k,j,i) = tend(k,j,i) + &
858	particles(n)%weight_factor / &
859	prt_count(k,j,i)
860	ENDIF
861	ENDDO
862	ENDDO
863	ENDDO
864	ENDDO
865	! CALL exchange_horiz( tend, nbgp )
866	ELSE
867	tend = 0.0_wp
868	ENDIF
869	DO i = nxl, nxr
870	DO j = nys, nyn
871	DO k = nzb, nzt+1
872	local_pf(i,j,k) = tend(k,j,i)
873	ENDDO
874	ENDDO
875	ENDDO
876	resorted = .TRUE.
877	ELSE
878	IF ( .NOT. ALLOCATED( ql_vp_av ) ) THEN
879	ALLOCATE( ql_vp_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
880	ql_vp_av = REAL( fill_value, KIND = wp )
881	ENDIF
882	! CALL exchange_horiz( ql_vp_av, nbgp )
883	to_be_resorted => ql_vp_av
884	ENDIF
885	IF ( mode == 'xy' ) level_z = zu
886
887	CASE ( 'qsws*_xy' ) ! 2d-array
888	IF ( av == 0 ) THEN
889	local_pf(:,:,nzb+1) = REAL( fill_value, KIND = wp )
890	!
891	!-- In case of default surfaces, clean-up flux by density.
892	!-- In case of land-surfaces, convert fluxes into
893	!-- dynamic units
894	DO m = 1, surf_def_h(0)%ns
895	i = surf_def_h(0)%i(m)
896	j = surf_def_h(0)%j(m)
897	k = surf_def_h(0)%k(m)
898	local_pf(i,j,nzb+1) = surf_def_h(0)%qsws(m) * &
899	waterflux_output_conversion(k)
900	ENDDO
901	DO m = 1, surf_lsm_h%ns
902	i = surf_lsm_h%i(m)
903	j = surf_lsm_h%j(m)
904	k = surf_lsm_h%k(m)
905	local_pf(i,j,nzb+1) = surf_lsm_h%qsws(m) * l_v
906	ENDDO
907	ELSE
908	IF ( .NOT. ALLOCATED( qsws_av ) ) THEN
909	ALLOCATE( qsws_av(nysg:nyng,nxlg:nxrg) )
910	qsws_av = REAL( fill_value, KIND = wp )
911	ENDIF
912	DO i = nxl, nxr
913	DO j = nys, nyn
914	local_pf(i,j,nzb+1) = qsws_av(j,i)
915	ENDDO
916	ENDDO
917	ENDIF
918	resorted = .TRUE.
919	two_d = .TRUE.
920	level_z(nzb+1) = zu(nzb+1)
921
922	CASE ( 'qv_xy', 'qv_xz', 'qv_yz' )
923	IF ( av == 0 ) THEN
924	DO i = nxl, nxr
925	DO j = nys, nyn
926	DO k = nzb, nzt+1
927	local_pf(i,j,k) = q(k,j,i) - ql(k,j,i)
928	ENDDO
929	ENDDO
930	ENDDO
931	resorted = .TRUE.
932	ELSE
933	IF ( .NOT. ALLOCATED( qv_av ) ) THEN
934	ALLOCATE( qv_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
935	qv_av = REAL( fill_value, KIND = wp )
936	ENDIF
937	to_be_resorted => qv_av
938	ENDIF
939	IF ( mode == 'xy' ) level_z = zu
940
941	CASE ( 'r_a*_xy' ) ! 2d-array
942	IF ( av == 0 ) THEN
943	DO m = 1, surf_lsm_h%ns
944	i = surf_lsm_h%i(m)
945	j = surf_lsm_h%j(m)
946	local_pf(i,j,nzb+1) = surf_lsm_h%r_a(m)
947	ENDDO
948
949	DO m = 1, surf_usm_h%ns
950	i = surf_usm_h%i(m)
951	j = surf_usm_h%j(m)
952	local_pf(i,j,nzb+1) = &
953	( surf_usm_h%frac(ind_veg_wall,m) * &
954	surf_usm_h%r_a(m) + &
955	surf_usm_h%frac(ind_pav_green,m) * &
956	surf_usm_h%r_a_green(m) + &
957	surf_usm_h%frac(ind_wat_win,m) * &
958	surf_usm_h%r_a_window(m) )
959	ENDDO
960	ELSE
961	IF ( .NOT. ALLOCATED( r_a_av ) ) THEN
962	ALLOCATE( r_a_av(nysg:nyng,nxlg:nxrg) )
963	r_a_av = REAL( fill_value, KIND = wp )
964	ENDIF
965	DO i = nxl, nxr
966	DO j = nys, nyn
967	local_pf(i,j,nzb+1) = r_a_av(j,i)
968	ENDDO
969	ENDDO
970	ENDIF
971	resorted = .TRUE.
972	two_d = .TRUE.
973	level_z(nzb+1) = zu(nzb+1)
974
975	CASE ( 's_xy', 's_xz', 's_yz' )
976	IF ( av == 0 ) THEN
977	to_be_resorted => s
978	ELSE
979	IF ( .NOT. ALLOCATED( s_av ) ) THEN
980	ALLOCATE( s_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
981	s_av = REAL( fill_value, KIND = wp )
982	ENDIF
983	to_be_resorted => s_av
984	ENDIF
985
986	CASE ( 'shf*_xy' ) ! 2d-array
987	IF ( av == 0 ) THEN
988	!
989	!-- In case of default surfaces, clean-up flux by density.
990	!-- In case of land- and urban-surfaces, convert fluxes into
991	!-- dynamic units.
992	DO m = 1, surf_def_h(0)%ns
993	i = surf_def_h(0)%i(m)
994	j = surf_def_h(0)%j(m)
995	k = surf_def_h(0)%k(m)
996	local_pf(i,j,nzb+1) = surf_def_h(0)%shf(m) * &
997	heatflux_output_conversion(k)
998	ENDDO
999	DO m = 1, surf_lsm_h%ns
1000	i = surf_lsm_h%i(m)
1001	j = surf_lsm_h%j(m)
1002	k = surf_lsm_h%k(m)
1003	local_pf(i,j,nzb+1) = surf_lsm_h%shf(m) * c_p
1004	ENDDO
1005	DO m = 1, surf_usm_h%ns
1006	i = surf_usm_h%i(m)
1007	j = surf_usm_h%j(m)
1008	k = surf_usm_h%k(m)
1009	local_pf(i,j,nzb+1) = surf_usm_h%shf(m) * c_p
1010	ENDDO
1011	ELSE
1012	IF ( .NOT. ALLOCATED( shf_av ) ) THEN
1013	ALLOCATE( shf_av(nysg:nyng,nxlg:nxrg) )
1014	shf_av = REAL( fill_value, KIND = wp )
1015	ENDIF
1016	DO i = nxl, nxr
1017	DO j = nys, nyn
1018	local_pf(i,j,nzb+1) = shf_av(j,i)
1019	ENDDO
1020	ENDDO
1021	ENDIF
1022	resorted = .TRUE.
1023	two_d = .TRUE.
1024	level_z(nzb+1) = zu(nzb+1)
1025
1026	CASE ( 'ssws*_xy' ) ! 2d-array
1027	IF ( av == 0 ) THEN
1028	DO m = 1, surf_def_h(0)%ns
1029	i = surf_def_h(0)%i(m)
1030	j = surf_def_h(0)%j(m)
1031	local_pf(i,j,nzb+1) = surf_def_h(0)%ssws(m)
1032	ENDDO
1033	DO m = 1, surf_lsm_h%ns
1034	i = surf_lsm_h%i(m)
1035	j = surf_lsm_h%j(m)
1036	local_pf(i,j,nzb+1) = surf_lsm_h%ssws(m)
1037	ENDDO
1038	DO m = 1, surf_usm_h%ns
1039	i = surf_usm_h%i(m)
1040	j = surf_usm_h%j(m)
1041	local_pf(i,j,nzb+1) = surf_usm_h%ssws(m)
1042	ENDDO
1043	ELSE
1044	IF ( .NOT. ALLOCATED( ssws_av ) ) THEN
1045	ALLOCATE( ssws_av(nysg:nyng,nxlg:nxrg) )
1046	ssws_av = REAL( fill_value, KIND = wp )
1047	ENDIF
1048	DO i = nxl, nxr
1049	DO j = nys, nyn
1050	local_pf(i,j,nzb+1) = ssws_av(j,i)
1051	ENDDO
1052	ENDDO
1053	ENDIF
1054	resorted = .TRUE.
1055	two_d = .TRUE.
1056	level_z(nzb+1) = zu(nzb+1)
1057
1058	CASE ( 't*_xy' ) ! 2d-array
1059	IF ( av == 0 ) THEN
1060	DO m = 1, surf_def_h(0)%ns
1061	i = surf_def_h(0)%i(m)
1062	j = surf_def_h(0)%j(m)
1063	local_pf(i,j,nzb+1) = surf_def_h(0)%ts(m)
1064	ENDDO
1065	DO m = 1, surf_lsm_h%ns
1066	i = surf_lsm_h%i(m)
1067	j = surf_lsm_h%j(m)
1068	local_pf(i,j,nzb+1) = surf_lsm_h%ts(m)
1069	ENDDO
1070	DO m = 1, surf_usm_h%ns
1071	i = surf_usm_h%i(m)
1072	j = surf_usm_h%j(m)
1073	local_pf(i,j,nzb+1) = surf_usm_h%ts(m)
1074	ENDDO
1075	ELSE
1076	IF ( .NOT. ALLOCATED( ts_av ) ) THEN
1077	ALLOCATE( ts_av(nysg:nyng,nxlg:nxrg) )
1078	ts_av = REAL( fill_value, KIND = wp )
1079	ENDIF
1080	DO i = nxl, nxr
1081	DO j = nys, nyn
1082	local_pf(i,j,nzb+1) = ts_av(j,i)
1083	ENDDO
1084	ENDDO
1085	ENDIF
1086	resorted = .TRUE.
1087	two_d = .TRUE.
1088	level_z(nzb+1) = zu(nzb+1)
1089
1090	CASE ( 'tsurf*_xy' ) ! 2d-array
1091	IF ( av == 0 ) THEN
1092	DO m = 1, surf_def_h(0)%ns
1093	i = surf_def_h(0)%i(m)
1094	j = surf_def_h(0)%j(m)
1095	local_pf(i,j,nzb+1) = surf_def_h(0)%pt_surface(m)
1096	ENDDO
1097
1098	DO m = 1, surf_lsm_h%ns
1099	i = surf_lsm_h%i(m)
1100	j = surf_lsm_h%j(m)
1101	local_pf(i,j,nzb+1) = surf_lsm_h%pt_surface(m)
1102	ENDDO
1103
1104	DO m = 1, surf_usm_h%ns
1105	i = surf_usm_h%i(m)
1106	j = surf_usm_h%j(m)
1107	local_pf(i,j,nzb+1) = surf_usm_h%pt_surface(m)
1108	ENDDO
1109
1110	ELSE
1111	IF ( .NOT. ALLOCATED( tsurf_av ) ) THEN
1112	ALLOCATE( tsurf_av(nysg:nyng,nxlg:nxrg) )
1113	tsurf_av = REAL( fill_value, KIND = wp )
1114	ENDIF
1115	DO i = nxl, nxr
1116	DO j = nys, nyn
1117	local_pf(i,j,nzb+1) = tsurf_av(j,i)
1118	ENDDO
1119	ENDDO
1120	ENDIF
1121	resorted = .TRUE.
1122	two_d = .TRUE.
1123	level_z(nzb+1) = zu(nzb+1)
1124
1125	CASE ( 'u_xy', 'u_xz', 'u_yz' )
1126	flag_nr = 1
1127	IF ( av == 0 ) THEN
1128	to_be_resorted => u
1129	ELSE
1130	IF ( .NOT. ALLOCATED( u_av ) ) THEN
1131	ALLOCATE( u_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1132	u_av = REAL( fill_value, KIND = wp )
1133	ENDIF
1134	to_be_resorted => u_av
1135	ENDIF
1136	IF ( mode == 'xy' ) level_z = zu
1137	!
1138	!-- Substitute the values generated by "mirror" boundary condition
1139	!-- at the bottom boundary by the real surface values.
1140	IF ( do2d(av,ivar) == 'u_xz' .OR. do2d(av,ivar) == 'u_yz' ) THEN
1141	IF ( ibc_uv_b == 0 ) local_pf(:,:,nzb) = 0.0_wp
1142	ENDIF
1143
1144	CASE ( 'us*_xy' ) ! 2d-array
1145	IF ( av == 0 ) THEN
1146	DO m = 1, surf_def_h(0)%ns
1147	i = surf_def_h(0)%i(m)
1148	j = surf_def_h(0)%j(m)
1149	local_pf(i,j,nzb+1) = surf_def_h(0)%us(m)
1150	ENDDO
1151	DO m = 1, surf_lsm_h%ns
1152	i = surf_lsm_h%i(m)
1153	j = surf_lsm_h%j(m)
1154	local_pf(i,j,nzb+1) = surf_lsm_h%us(m)
1155	ENDDO
1156	DO m = 1, surf_usm_h%ns
1157	i = surf_usm_h%i(m)
1158	j = surf_usm_h%j(m)
1159	local_pf(i,j,nzb+1) = surf_usm_h%us(m)
1160	ENDDO
1161	ELSE
1162	IF ( .NOT. ALLOCATED( us_av ) ) THEN
1163	ALLOCATE( us_av(nysg:nyng,nxlg:nxrg) )
1164	us_av = REAL( fill_value, KIND = wp )
1165	ENDIF
1166	DO i = nxl, nxr
1167	DO j = nys, nyn
1168	local_pf(i,j,nzb+1) = us_av(j,i)
1169	ENDDO
1170	ENDDO
1171	ENDIF
1172	resorted = .TRUE.
1173	two_d = .TRUE.
1174	level_z(nzb+1) = zu(nzb+1)
1175
1176	CASE ( 'v_xy', 'v_xz', 'v_yz' )
1177	flag_nr = 2
1178	IF ( av == 0 ) THEN
1179	to_be_resorted => v
1180	ELSE
1181	IF ( .NOT. ALLOCATED( v_av ) ) THEN
1182	ALLOCATE( v_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1183	v_av = REAL( fill_value, KIND = wp )
1184	ENDIF
1185	to_be_resorted => v_av
1186	ENDIF
1187	IF ( mode == 'xy' ) level_z = zu
1188	!
1189	!-- Substitute the values generated by "mirror" boundary condition
1190	!-- at the bottom boundary by the real surface values.
1191	IF ( do2d(av,ivar) == 'v_xz' .OR. do2d(av,ivar) == 'v_yz' ) THEN
1192	IF ( ibc_uv_b == 0 ) local_pf(:,:,nzb) = 0.0_wp
1193	ENDIF
1194
1195	CASE ( 'thetav_xy', 'thetav_xz', 'thetav_yz' )
1196	IF ( av == 0 ) THEN
1197	to_be_resorted => vpt
1198	ELSE
1199	IF ( .NOT. ALLOCATED( vpt_av ) ) THEN
1200	ALLOCATE( vpt_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1201	vpt_av = REAL( fill_value, KIND = wp )
1202	ENDIF
1203	to_be_resorted => vpt_av
1204	ENDIF
1205	IF ( mode == 'xy' ) level_z = zu
1206
1207	CASE ( 'w_xy', 'w_xz', 'w_yz' )
1208	flag_nr = 3
1209	IF ( av == 0 ) THEN
1210	to_be_resorted => w
1211	ELSE
1212	IF ( .NOT. ALLOCATED( w_av ) ) THEN
1213	ALLOCATE( w_av(nzb:nzt+1,nysg:nyng,nxlg:nxrg) )
1214	w_av = REAL( fill_value, KIND = wp )
1215	ENDIF
1216	to_be_resorted => w_av
1217	ENDIF
1218	IF ( mode == 'xy' ) level_z = zw
1219
1220	CASE ( 'z0*_xy' ) ! 2d-array
1221	IF ( av == 0 ) THEN
1222	DO m = 1, surf_def_h(0)%ns
1223	i = surf_def_h(0)%i(m)
1224	j = surf_def_h(0)%j(m)
1225	local_pf(i,j,nzb+1) = surf_def_h(0)%z0(m)
1226	ENDDO
1227	DO m = 1, surf_lsm_h%ns
1228	i = surf_lsm_h%i(m)
1229	j = surf_lsm_h%j(m)
1230	local_pf(i,j,nzb+1) = surf_lsm_h%z0(m)
1231	ENDDO
1232	DO m = 1, surf_usm_h%ns
1233	i = surf_usm_h%i(m)
1234	j = surf_usm_h%j(m)
1235	local_pf(i,j,nzb+1) = surf_usm_h%z0(m)
1236	ENDDO
1237	ELSE
1238	IF ( .NOT. ALLOCATED( z0_av ) ) THEN
1239	ALLOCATE( z0_av(nysg:nyng,nxlg:nxrg) )
1240	z0_av = REAL( fill_value, KIND = wp )
1241	ENDIF
1242	DO i = nxl, nxr
1243	DO j = nys, nyn
1244	local_pf(i,j,nzb+1) = z0_av(j,i)
1245	ENDDO
1246	ENDDO
1247	ENDIF
1248	resorted = .TRUE.
1249	two_d = .TRUE.
1250	level_z(nzb+1) = zu(nzb+1)
1251
1252	CASE ( 'z0h*_xy' ) ! 2d-array
1253	IF ( av == 0 ) THEN
1254	DO m = 1, surf_def_h(0)%ns
1255	i = surf_def_h(0)%i(m)
1256	j = surf_def_h(0)%j(m)
1257	local_pf(i,j,nzb+1) = surf_def_h(0)%z0h(m)
1258	ENDDO
1259	DO m = 1, surf_lsm_h%ns
1260	i = surf_lsm_h%i(m)
1261	j = surf_lsm_h%j(m)
1262	local_pf(i,j,nzb+1) = surf_lsm_h%z0h(m)
1263	ENDDO
1264	DO m = 1, surf_usm_h%ns
1265	i = surf_usm_h%i(m)
1266	j = surf_usm_h%j(m)
1267	local_pf(i,j,nzb+1) = surf_usm_h%z0h(m)
1268	ENDDO
1269	ELSE
1270	IF ( .NOT. ALLOCATED( z0h_av ) ) THEN
1271	ALLOCATE( z0h_av(nysg:nyng,nxlg:nxrg) )
1272	z0h_av = REAL( fill_value, KIND = wp )
1273	ENDIF
1274	DO i = nxl, nxr
1275	DO j = nys, nyn
1276	local_pf(i,j,nzb+1) = z0h_av(j,i)
1277	ENDDO
1278	ENDDO
1279	ENDIF
1280	resorted = .TRUE.
1281	two_d = .TRUE.
1282	level_z(nzb+1) = zu(nzb+1)
1283
1284	CASE ( 'z0q*_xy' ) ! 2d-array
1285	IF ( av == 0 ) THEN
1286	DO m = 1, surf_def_h(0)%ns
1287	i = surf_def_h(0)%i(m)
1288	j = surf_def_h(0)%j(m)
1289	local_pf(i,j,nzb+1) = surf_def_h(0)%z0q(m)
1290	ENDDO
1291	DO m = 1, surf_lsm_h%ns
1292	i = surf_lsm_h%i(m)
1293	j = surf_lsm_h%j(m)
1294	local_pf(i,j,nzb+1) = surf_lsm_h%z0q(m)
1295	ENDDO
1296	DO m = 1, surf_usm_h%ns
1297	i = surf_usm_h%i(m)
1298	j = surf_usm_h%j(m)
1299	local_pf(i,j,nzb+1) = surf_usm_h%z0q(m)
1300	ENDDO
1301	ELSE
1302	IF ( .NOT. ALLOCATED( z0q_av ) ) THEN
1303	ALLOCATE( z0q_av(nysg:nyng,nxlg:nxrg) )
1304	z0q_av = REAL( fill_value, KIND = wp )
1305	ENDIF
1306	DO i = nxl, nxr
1307	DO j = nys, nyn
1308	local_pf(i,j,nzb+1) = z0q_av(j,i)
1309	ENDDO
1310	ENDDO
1311	ENDIF
1312	resorted = .TRUE.
1313	two_d = .TRUE.
1314	level_z(nzb+1) = zu(nzb+1)
1315
1316	CASE DEFAULT
1317
1318	!
1319	!-- Quantities of other modules
1320	IF ( .NOT. found .AND. bulk_cloud_model ) THEN
1321	CALL bcm_data_output_2d( av, do2d(av,ivar), found, grid, mode,&
1322	local_pf, two_d, nzb_do, nzt_do )
1323	ENDIF
1324
1325	IF ( .NOT. found .AND. gust_module_enabled ) THEN
1326	CALL gust_data_output_2d( av, do2d(av,ivar), found, grid, &
1327	local_pf, two_d, nzb_do, nzt_do )
1328	ENDIF
1329
1330	IF ( .NOT. found .AND. biometeorology &
1331	.AND. mode == 'xy' ) THEN
1332	CALL bio_data_output_2d( av, do2d(av,ivar), found, grid, &
1333	local_pf, two_d, nzb_do, nzt_do, &
1334	fill_value )
1335	ENDIF
1336
1337	IF ( .NOT. found .AND. land_surface ) THEN
1338	CALL lsm_data_output_2d( av, do2d(av,ivar), found, grid, mode,&
1339	local_pf, two_d, nzb_do, nzt_do )
1340	ENDIF
1341
1342	IF ( .NOT. found .AND. ocean_mode ) THEN
1343	CALL ocean_data_output_2d( av, do2d(av,ivar), found, grid, &
1344	mode, local_pf, nzb_do, nzt_do )
1345	ENDIF
1346
1347	IF ( .NOT. found .AND. radiation ) THEN
1348	CALL radiation_data_output_2d( av, do2d(av,ivar), found, grid,&
1349	mode, local_pf, two_d, &
1350	nzb_do, nzt_do )
1351	ENDIF
1352
1353	IF ( .NOT. found .AND. salsa ) THEN
1354	CALL salsa_data_output_2d( av, do2d(av,ivar), found, grid, &
1355	mode, local_pf, two_d )
1356	ENDIF
1357
1358	IF ( .NOT. found .AND. uv_exposure ) THEN
1359	CALL uvem_data_output_2d( av, do2d(av,ivar), found, grid, &
1360	local_pf, two_d, nzb_do, nzt_do )
1361	ENDIF
1362
1363	IF ( .NOT. found .AND. air_chemistry ) THEN
1364	CALL chem_data_output_2d( av, do2d(av,ivar), found, grid, mode, &
1365	local_pf, two_d, nzb_do, nzt_do, fill_value )
1366	ENDIF
1367	!
1368	!-- User defined quantities
1369	IF ( .NOT. found ) THEN
1370	CALL user_data_output_2d( av, do2d(av,ivar), found, grid, &
1371	local_pf, two_d, nzb_do, nzt_do )
1372	ENDIF
1373
1374	resorted = .TRUE.
1375
1376	IF ( grid == 'zu' ) THEN
1377	IF ( mode == 'xy' ) level_z = zu
1378	ELSEIF ( grid == 'zw' ) THEN
1379	IF ( mode == 'xy' ) level_z = zw
1380	ELSEIF ( grid == 'zu1' ) THEN
1381	IF ( mode == 'xy' ) level_z(nzb+1) = zu(nzb+1)
1382	ELSEIF ( grid == 'zs' ) THEN
1383	IF ( mode == 'xy' ) level_z = zs
1384	ENDIF
1385
1386	IF ( .NOT. found ) THEN
1387	message_string = 'no output provided for: ' // &
1388	TRIM( do2d(av,ivar) )
1389	CALL message( 'data_output_2d', 'PA0181', 0, 0, 0, 6, 0 )
1390	ENDIF
1391
1392	END SELECT
1393
1394	!
1395	!-- Resort the array to be output, if not done above. Flag topography
1396	!-- grid points with fill values, using the corresponding maksing flag.
1397	IF ( .NOT. resorted ) THEN
1398	DO i = nxl, nxr
1399	DO j = nys, nyn
1400	DO k = nzb_do, nzt_do
1401	local_pf(i,j,k) = MERGE( to_be_resorted(k,j,i), &
1402	REAL( fill_value, KIND = wp ), &
1403	BTEST( wall_flags_0(k,j,i), &
1404	flag_nr ) )
1405	ENDDO
1406	ENDDO
1407	ENDDO
1408	ENDIF
1409
1410	!
1411	!-- Output of the individual cross-sections, depending on the cross-
1412	!-- section mode chosen.
1413	is = 1
1414	loop1: DO WHILE ( section(is,s_ind) /= -9999 .OR. two_d )
1415
1416	SELECT CASE ( mode )
1417
1418	CASE ( 'xy' )
1419	!
1420	!-- Determine the cross section index
1421	IF ( two_d ) THEN
1422	layer_xy = nzb+1
1423	ELSE
1424	layer_xy = section(is,s_ind)
1425	ENDIF
1426
1427	!
1428	!-- Exit the loop for layers beyond the data output domain
1429	!-- (used for soil model)
1430	IF ( layer_xy > nzt_do ) THEN
1431	EXIT loop1
1432	ENDIF
1433
1434	!
1435	!-- Update the netCDF xy cross section time axis.
1436	!-- In case of parallel output, this is only done by PE0
1437	!-- to increase the performance.
1438	IF ( simulated_time /= do2d_xy_last_time(av) ) THEN
1439	do2d_xy_time_count(av) = do2d_xy_time_count(av) + 1
1440	do2d_xy_last_time(av) = simulated_time
1441	IF ( myid == 0 ) THEN
1442	IF ( .NOT. data_output_2d_on_each_pe &
1443	.OR. netcdf_data_format > 4 ) &
1444	THEN
1445	#if defined( __netcdf )
1446	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1447	id_var_time_xy(av), &
1448	(/ time_since_reference_point /), &
1449	start = (/ do2d_xy_time_count(av) /), &
1450	count = (/ 1 /) )
1451	CALL netcdf_handle_error( 'data_output_2d', 53 )
1452	#endif
1453	ENDIF
1454	ENDIF
1455	ENDIF
1456	!
1457	!-- If required, carry out averaging along z
1458	IF ( section(is,s_ind) == -1 .AND. .NOT. two_d ) THEN
1459
1460	local_2d = 0.0_wp
1461	!
1462	!-- Carry out the averaging (all data are on the PE)
1463	DO k = nzb_do, nzt_do
1464	DO j = nys, nyn
1465	DO i = nxl, nxr
1466	local_2d(i,j) = local_2d(i,j) + local_pf(i,j,k)
1467	ENDDO
1468	ENDDO
1469	ENDDO
1470
1471	local_2d = local_2d / ( nzt_do - nzb_do + 1.0_wp)
1472
1473	ELSE
1474	!
1475	!-- Just store the respective section on the local array
1476	local_2d = local_pf(:,:,layer_xy)
1477
1478	ENDIF
1479
1480	#if defined( __parallel )
1481	IF ( netcdf_data_format > 4 ) THEN
1482	!
1483	!-- Parallel output in netCDF4/HDF5 format.
1484	IF ( two_d ) THEN
1485	iis = 1
1486	ELSE
1487	iis = is
1488	ENDIF
1489
1490	#if defined( __netcdf )
1491	!
1492	!-- For parallel output, all cross sections are first stored
1493	!-- here on a local array and will be written to the output
1494	!-- file afterwards to increase the performance.
1495	DO i = nxl, nxr
1496	DO j = nys, nyn
1497	local_2d_sections(i,j,iis) = local_2d(i,j)
1498	ENDDO
1499	ENDDO
1500	#endif
1501	ELSE
1502
1503	IF ( data_output_2d_on_each_pe ) THEN
1504	!
1505	!-- Output of partial arrays on each PE
1506	#if defined( __netcdf )
1507	IF ( myid == 0 ) THEN
1508	WRITE ( 21 ) time_since_reference_point, &
1509	do2d_xy_time_count(av), av
1510	ENDIF
1511	#endif
1512	DO i = 0, io_blocks-1
1513	IF ( i == io_group ) THEN
1514	WRITE ( 21 ) nxl, nxr, nys, nyn, nys, nyn
1515	WRITE ( 21 ) local_2d
1516	ENDIF
1517	#if defined( __parallel )
1518	CALL MPI_BARRIER( comm2d, ierr )
1519	#endif
1520	ENDDO
1521
1522	ELSE
1523	!
1524	!-- PE0 receives partial arrays from all processors and
1525	!-- then outputs them. Here a barrier has to be set,
1526	!-- because otherwise "-MPI- FATAL: Remote protocol queue
1527	!-- full" may occur.
1528	CALL MPI_BARRIER( comm2d, ierr )
1529
1530	ngp = ( nxr-nxl+1 ) * ( nyn-nys+1 )
1531	IF ( myid == 0 ) THEN
1532	!
1533	!-- Local array can be relocated directly.
1534	total_2d(nxl:nxr,nys:nyn) = local_2d
1535	!
1536	!-- Receive data from all other PEs.
1537	DO n = 1, numprocs-1
1538	!
1539	!-- Receive index limits first, then array.
1540	!-- Index limits are received in arbitrary order from
1541	!-- the PEs.
1542	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1543	MPI_ANY_SOURCE, 0, comm2d, &
1544	status, ierr )
1545	sender = status(MPI_SOURCE)
1546	DEALLOCATE( local_2d )
1547	ALLOCATE( local_2d(ind(1):ind(2),ind(3):ind(4)) )
1548	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
1549	MPI_REAL, sender, 1, comm2d, &
1550	status, ierr )
1551	total_2d(ind(1):ind(2),ind(3):ind(4)) = local_2d
1552	ENDDO
1553	!
1554	!-- Relocate the local array for the next loop increment
1555	DEALLOCATE( local_2d )
1556	ALLOCATE( local_2d(nxl:nxr,nys:nyn) )
1557
1558	#if defined( __netcdf )
1559	IF ( two_d ) THEN
1560	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1561	id_var_do2d(av,ivar), &
1562	total_2d(0:nx,0:ny), &
1563	start = (/ 1, 1, 1, do2d_xy_time_count(av) /), &
1564	count = (/ nx+1, ny+1, 1, 1 /) )
1565	ELSE
1566	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1567	id_var_do2d(av,ivar), &
1568	total_2d(0:nx,0:ny), &
1569	start = (/ 1, 1, is, do2d_xy_time_count(av) /), &
1570	count = (/ nx+1, ny+1, 1, 1 /) )
1571	ENDIF
1572	CALL netcdf_handle_error( 'data_output_2d', 54 )
1573	#endif
1574
1575	ELSE
1576	!
1577	!-- First send the local index limits to PE0
1578	ind(1) = nxl; ind(2) = nxr
1579	ind(3) = nys; ind(4) = nyn
1580	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
1581	comm2d, ierr )
1582	!
1583	!-- Send data to PE0
1584	CALL MPI_SEND( local_2d(nxl,nys), ngp, &
1585	MPI_REAL, 0, 1, comm2d, ierr )
1586	ENDIF
1587	!
1588	!-- A barrier has to be set, because otherwise some PEs may
1589	!-- proceed too fast so that PE0 may receive wrong data on
1590	!-- tag 0
1591	CALL MPI_BARRIER( comm2d, ierr )
1592	ENDIF
1593
1594	ENDIF
1595	#else
1596	#if defined( __netcdf )
1597	IF ( two_d ) THEN
1598	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1599	id_var_do2d(av,ivar), &
1600	local_2d(nxl:nxr,nys:nyn), &
1601	start = (/ 1, 1, 1, do2d_xy_time_count(av) /), &
1602	count = (/ nx+1, ny+1, 1, 1 /) )
1603	ELSE
1604	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
1605	id_var_do2d(av,ivar), &
1606	local_2d(nxl:nxr,nys:nyn), &
1607	start = (/ 1, 1, is, do2d_xy_time_count(av) /), &
1608	count = (/ nx+1, ny+1, 1, 1 /) )
1609	ENDIF
1610	CALL netcdf_handle_error( 'data_output_2d', 447 )
1611	#endif
1612	#endif
1613
1614	!
1615	!-- For 2D-arrays (e.g. u*) only one cross-section is available.
1616	!-- Hence exit loop of output levels.
1617	IF ( two_d ) THEN
1618	IF ( netcdf_data_format < 5 ) two_d = .FALSE.
1619	EXIT loop1
1620	ENDIF
1621
1622	CASE ( 'xz' )
1623	!
1624	!-- Update the netCDF xz cross section time axis.
1625	!-- In case of parallel output, this is only done by PE0
1626	!-- to increase the performance.
1627	IF ( simulated_time /= do2d_xz_last_time(av) ) THEN
1628	do2d_xz_time_count(av) = do2d_xz_time_count(av) + 1
1629	do2d_xz_last_time(av) = simulated_time
1630	IF ( myid == 0 ) THEN
1631	IF ( .NOT. data_output_2d_on_each_pe &
1632	.OR. netcdf_data_format > 4 ) &
1633	THEN
1634	#if defined( __netcdf )
1635	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1636	id_var_time_xz(av), &
1637	(/ time_since_reference_point /), &
1638	start = (/ do2d_xz_time_count(av) /), &
1639	count = (/ 1 /) )
1640	CALL netcdf_handle_error( 'data_output_2d', 56 )
1641	#endif
1642	ENDIF
1643	ENDIF
1644	ENDIF
1645
1646	!
1647	!-- If required, carry out averaging along y
1648	IF ( section(is,s_ind) == -1 ) THEN
1649
1650	ALLOCATE( local_2d_l(nxl:nxr,nzb_do:nzt_do) )
1651	local_2d_l = 0.0_wp
1652	ngp = ( nxr-nxl + 1 ) * ( nzt_do-nzb_do + 1 )
1653	!
1654	!-- First local averaging on the PE
1655	DO k = nzb_do, nzt_do
1656	DO j = nys, nyn
1657	DO i = nxl, nxr
1658	local_2d_l(i,k) = local_2d_l(i,k) + &
1659	local_pf(i,j,k)
1660	ENDDO
1661	ENDDO
1662	ENDDO
1663	#if defined( __parallel )
1664	!
1665	!-- Now do the averaging over all PEs along y
1666	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
1667	CALL MPI_ALLREDUCE( local_2d_l(nxl,nzb_do), &
1668	local_2d(nxl,nzb_do), ngp, MPI_REAL, &
1669	MPI_SUM, comm1dy, ierr )
1670	#else
1671	local_2d = local_2d_l
1672	#endif
1673	local_2d = local_2d / ( ny + 1.0_wp )
1674
1675	DEALLOCATE( local_2d_l )
1676
1677	ELSE
1678	!
1679	!-- Just store the respective section on the local array
1680	!-- (but only if it is available on this PE!)
1681	IF ( section(is,s_ind) >= nys .AND. section(is,s_ind) <= nyn ) &
1682	THEN
1683	local_2d = local_pf(:,section(is,s_ind),nzb_do:nzt_do)
1684	ENDIF
1685
1686	ENDIF
1687
1688	#if defined( __parallel )
1689	IF ( netcdf_data_format > 4 ) THEN
1690	!
1691	!-- Output in netCDF4/HDF5 format.
1692	!-- Output only on those PEs where the respective cross
1693	!-- sections reside. Cross sections averaged along y are
1694	!-- output on the respective first PE along y (myidy=0).
1695	IF ( ( section(is,s_ind) >= nys .AND. &
1696	section(is,s_ind) <= nyn ) .OR. &
1697	( section(is,s_ind) == -1 .AND. myidy == 0 ) ) THEN
1698	#if defined( __netcdf )
1699	!
1700	!-- For parallel output, all cross sections are first
1701	!-- stored here on a local array and will be written to the
1702	!-- output file afterwards to increase the performance.
1703	DO i = nxl, nxr
1704	DO k = nzb_do, nzt_do
1705	local_2d_sections_l(i,is,k) = local_2d(i,k)
1706	ENDDO
1707	ENDDO
1708	#endif
1709	ENDIF
1710
1711	ELSE
1712
1713	IF ( data_output_2d_on_each_pe ) THEN
1714	!
1715	!-- Output of partial arrays on each PE. If the cross
1716	!-- section does not reside on the PE, output special
1717	!-- index values.
1718	#if defined( __netcdf )
1719	IF ( myid == 0 ) THEN
1720	WRITE ( 22 ) time_since_reference_point, &
1721	do2d_xz_time_count(av), av
1722	ENDIF
1723	#endif
1724	DO i = 0, io_blocks-1
1725	IF ( i == io_group ) THEN
1726	IF ( ( section(is,s_ind) >= nys .AND. &
1727	section(is,s_ind) <= nyn ) .OR. &
1728	( section(is,s_ind) == -1 .AND. &
1729	nys-1 == -1 ) ) &
1730	THEN
1731	WRITE (22) nxl, nxr, nzb_do, nzt_do, nzb, nzt+1
1732	WRITE (22) local_2d
1733	ELSE
1734	WRITE (22) -1, -1, -1, -1, -1, -1
1735	ENDIF
1736	ENDIF
1737	#if defined( __parallel )
1738	CALL MPI_BARRIER( comm2d, ierr )
1739	#endif
1740	ENDDO
1741
1742	ELSE
1743	!
1744	!-- PE0 receives partial arrays from all processors of the
1745	!-- respective cross section and outputs them. Here a
1746	!-- barrier has to be set, because otherwise
1747	!-- "-MPI- FATAL: Remote protocol queue full" may occur.
1748	CALL MPI_BARRIER( comm2d, ierr )
1749
1750	ngp = ( nxr-nxl + 1 ) * ( nzt_do-nzb_do + 1 )
1751	IF ( myid == 0 ) THEN
1752	!
1753	!-- Local array can be relocated directly.
1754	IF ( ( section(is,s_ind) >= nys .AND. &
1755	section(is,s_ind) <= nyn ) .OR. &
1756	( section(is,s_ind) == -1 .AND. &
1757	nys-1 == -1 ) ) THEN
1758	total_2d(nxl:nxr,nzb_do:nzt_do) = local_2d
1759	ENDIF
1760	!
1761	!-- Receive data from all other PEs.
1762	DO n = 1, numprocs-1
1763	!
1764	!-- Receive index limits first, then array.
1765	!-- Index limits are received in arbitrary order from
1766	!-- the PEs.
1767	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1768	MPI_ANY_SOURCE, 0, comm2d, &
1769	status, ierr )
1770	!
1771	!-- Not all PEs have data for XZ-cross-section.
1772	IF ( ind(1) /= -9999 ) THEN
1773	sender = status(MPI_SOURCE)
1774	DEALLOCATE( local_2d )
1775	ALLOCATE( local_2d(ind(1):ind(2), &
1776	ind(3):ind(4)) )
1777	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
1778	MPI_REAL, sender, 1, comm2d, &
1779	status, ierr )
1780	total_2d(ind(1):ind(2),ind(3):ind(4)) = &
1781	local_2d
1782	ENDIF
1783	ENDDO
1784	!
1785	!-- Relocate the local array for the next loop increment
1786	DEALLOCATE( local_2d )
1787	ALLOCATE( local_2d(nxl:nxr,nzb_do:nzt_do) )
1788
1789	#if defined( __netcdf )
1790	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1791	id_var_do2d(av,ivar), &
1792	total_2d(0:nx,nzb_do:nzt_do), &
1793	start = (/ 1, is, 1, do2d_xz_time_count(av) /), &
1794	count = (/ nx+1, 1, nzt_do-nzb_do+1, 1 /) )
1795	CALL netcdf_handle_error( 'data_output_2d', 58 )
1796	#endif
1797
1798	ELSE
1799	!
1800	!-- If the cross section resides on the PE, send the
1801	!-- local index limits, otherwise send -9999 to PE0.
1802	IF ( ( section(is,s_ind) >= nys .AND. &
1803	section(is,s_ind) <= nyn ) .OR. &
1804	( section(is,s_ind) == -1 .AND. nys-1 == -1 ) ) &
1805	THEN
1806	ind(1) = nxl; ind(2) = nxr
1807	ind(3) = nzb_do; ind(4) = nzt_do
1808	ELSE
1809	ind(1) = -9999; ind(2) = -9999
1810	ind(3) = -9999; ind(4) = -9999
1811	ENDIF
1812	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
1813	comm2d, ierr )
1814	!
1815	!-- If applicable, send data to PE0.
1816	IF ( ind(1) /= -9999 ) THEN
1817	CALL MPI_SEND( local_2d(nxl,nzb_do), ngp, &
1818	MPI_REAL, 0, 1, comm2d, ierr )
1819	ENDIF
1820	ENDIF
1821	!
1822	!-- A barrier has to be set, because otherwise some PEs may
1823	!-- proceed too fast so that PE0 may receive wrong data on
1824	!-- tag 0
1825	CALL MPI_BARRIER( comm2d, ierr )
1826	ENDIF
1827
1828	ENDIF
1829	#else
1830	#if defined( __netcdf )
1831	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
1832	id_var_do2d(av,ivar), &
1833	local_2d(nxl:nxr,nzb_do:nzt_do), &
1834	start = (/ 1, is, 1, do2d_xz_time_count(av) /), &
1835	count = (/ nx+1, 1, nzt_do-nzb_do+1, 1 /) )
1836	CALL netcdf_handle_error( 'data_output_2d', 451 )
1837	#endif
1838	#endif
1839
1840	CASE ( 'yz' )
1841	!
1842	!-- Update the netCDF yz cross section time axis.
1843	!-- In case of parallel output, this is only done by PE0
1844	!-- to increase the performance.
1845	IF ( simulated_time /= do2d_yz_last_time(av) ) THEN
1846	do2d_yz_time_count(av) = do2d_yz_time_count(av) + 1
1847	do2d_yz_last_time(av) = simulated_time
1848	IF ( myid == 0 ) THEN
1849	IF ( .NOT. data_output_2d_on_each_pe &
1850	.OR. netcdf_data_format > 4 ) &
1851	THEN
1852	#if defined( __netcdf )
1853	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
1854	id_var_time_yz(av), &
1855	(/ time_since_reference_point /), &
1856	start = (/ do2d_yz_time_count(av) /), &
1857	count = (/ 1 /) )
1858	CALL netcdf_handle_error( 'data_output_2d', 59 )
1859	#endif
1860	ENDIF
1861	ENDIF
1862	ENDIF
1863
1864	!
1865	!-- If required, carry out averaging along x
1866	IF ( section(is,s_ind) == -1 ) THEN
1867
1868	ALLOCATE( local_2d_l(nys:nyn,nzb_do:nzt_do) )
1869	local_2d_l = 0.0_wp
1870	ngp = ( nyn-nys+1 ) * ( nzt_do-nzb_do+1 )
1871	!
1872	!-- First local averaging on the PE
1873	DO k = nzb_do, nzt_do
1874	DO j = nys, nyn
1875	DO i = nxl, nxr
1876	local_2d_l(j,k) = local_2d_l(j,k) + &
1877	local_pf(i,j,k)
1878	ENDDO
1879	ENDDO
1880	ENDDO
1881	#if defined( __parallel )
1882	!
1883	!-- Now do the averaging over all PEs along x
1884	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
1885	CALL MPI_ALLREDUCE( local_2d_l(nys,nzb_do), &
1886	local_2d(nys,nzb_do), ngp, MPI_REAL, &
1887	MPI_SUM, comm1dx, ierr )
1888	#else
1889	local_2d = local_2d_l
1890	#endif
1891	local_2d = local_2d / ( nx + 1.0_wp )
1892
1893	DEALLOCATE( local_2d_l )
1894
1895	ELSE
1896	!
1897	!-- Just store the respective section on the local array
1898	!-- (but only if it is available on this PE!)
1899	IF ( section(is,s_ind) >= nxl .AND. section(is,s_ind) <= nxr ) &
1900	THEN
1901	local_2d = local_pf(section(is,s_ind),:,nzb_do:nzt_do)
1902	ENDIF
1903
1904	ENDIF
1905
1906	#if defined( __parallel )
1907	IF ( netcdf_data_format > 4 ) THEN
1908	!
1909	!-- Output in netCDF4/HDF5 format.
1910	!-- Output only on those PEs where the respective cross
1911	!-- sections reside. Cross sections averaged along x are
1912	!-- output on the respective first PE along x (myidx=0).
1913	IF ( ( section(is,s_ind) >= nxl .AND. &
1914	section(is,s_ind) <= nxr ) .OR. &
1915	( section(is,s_ind) == -1 .AND. myidx == 0 ) ) THEN
1916	#if defined( __netcdf )
1917	!
1918	!-- For parallel output, all cross sections are first
1919	!-- stored here on a local array and will be written to the
1920	!-- output file afterwards to increase the performance.
1921	DO j = nys, nyn
1922	DO k = nzb_do, nzt_do
1923	local_2d_sections_l(is,j,k) = local_2d(j,k)
1924	ENDDO
1925	ENDDO
1926	#endif
1927	ENDIF
1928
1929	ELSE
1930
1931	IF ( data_output_2d_on_each_pe ) THEN
1932	!
1933	!-- Output of partial arrays on each PE. If the cross
1934	!-- section does not reside on the PE, output special
1935	!-- index values.
1936	#if defined( __netcdf )
1937	IF ( myid == 0 ) THEN
1938	WRITE ( 23 ) time_since_reference_point, &
1939	do2d_yz_time_count(av), av
1940	ENDIF
1941	#endif
1942	DO i = 0, io_blocks-1
1943	IF ( i == io_group ) THEN
1944	IF ( ( section(is,s_ind) >= nxl .AND. &
1945	section(is,s_ind) <= nxr ) .OR. &
1946	( section(is,s_ind) == -1 .AND. &
1947	nxl-1 == -1 ) ) &
1948	THEN
1949	WRITE (23) nys, nyn, nzb_do, nzt_do, nzb, nzt+1
1950	WRITE (23) local_2d
1951	ELSE
1952	WRITE (23) -1, -1, -1, -1, -1, -1
1953	ENDIF
1954	ENDIF
1955	#if defined( __parallel )
1956	CALL MPI_BARRIER( comm2d, ierr )
1957	#endif
1958	ENDDO
1959
1960	ELSE
1961	!
1962	!-- PE0 receives partial arrays from all processors of the
1963	!-- respective cross section and outputs them. Here a
1964	!-- barrier has to be set, because otherwise
1965	!-- "-MPI- FATAL: Remote protocol queue full" may occur.
1966	CALL MPI_BARRIER( comm2d, ierr )
1967
1968	ngp = ( nyn-nys+1 ) * ( nzt_do-nzb_do+1 )
1969	IF ( myid == 0 ) THEN
1970	!
1971	!-- Local array can be relocated directly.
1972	IF ( ( section(is,s_ind) >= nxl .AND. &
1973	section(is,s_ind) <= nxr ) .OR. &
1974	( section(is,s_ind) == -1 .AND. nxl-1 == -1 ) ) &
1975	THEN
1976	total_2d(nys:nyn,nzb_do:nzt_do) = local_2d
1977	ENDIF
1978	!
1979	!-- Receive data from all other PEs.
1980	DO n = 1, numprocs-1
1981	!
1982	!-- Receive index limits first, then array.
1983	!-- Index limits are received in arbitrary order from
1984	!-- the PEs.
1985	CALL MPI_RECV( ind(1), 4, MPI_INTEGER, &
1986	MPI_ANY_SOURCE, 0, comm2d, &
1987	status, ierr )
1988	!
1989	!-- Not all PEs have data for YZ-cross-section.
1990	IF ( ind(1) /= -9999 ) THEN
1991	sender = status(MPI_SOURCE)
1992	DEALLOCATE( local_2d )
1993	ALLOCATE( local_2d(ind(1):ind(2), &
1994	ind(3):ind(4)) )
1995	CALL MPI_RECV( local_2d(ind(1),ind(3)), ngp, &
1996	MPI_REAL, sender, 1, comm2d, &
1997	status, ierr )
1998	total_2d(ind(1):ind(2),ind(3):ind(4)) = &
1999	local_2d
2000	ENDIF
2001	ENDDO
2002	!
2003	!-- Relocate the local array for the next loop increment
2004	DEALLOCATE( local_2d )
2005	ALLOCATE( local_2d(nys:nyn,nzb_do:nzt_do) )
2006
2007	#if defined( __netcdf )
2008	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2009	id_var_do2d(av,ivar), &
2010	total_2d(0:ny,nzb_do:nzt_do), &
2011	start = (/ is, 1, 1, do2d_yz_time_count(av) /), &
2012	count = (/ 1, ny+1, nzt_do-nzb_do+1, 1 /) )
2013	CALL netcdf_handle_error( 'data_output_2d', 61 )
2014	#endif
2015
2016	ELSE
2017	!
2018	!-- If the cross section resides on the PE, send the
2019	!-- local index limits, otherwise send -9999 to PE0.
2020	IF ( ( section(is,s_ind) >= nxl .AND. &
2021	section(is,s_ind) <= nxr ) .OR. &
2022	( section(is,s_ind) == -1 .AND. nxl-1 == -1 ) ) &
2023	THEN
2024	ind(1) = nys; ind(2) = nyn
2025	ind(3) = nzb_do; ind(4) = nzt_do
2026	ELSE
2027	ind(1) = -9999; ind(2) = -9999
2028	ind(3) = -9999; ind(4) = -9999
2029	ENDIF
2030	CALL MPI_SEND( ind(1), 4, MPI_INTEGER, 0, 0, &
2031	comm2d, ierr )
2032	!
2033	!-- If applicable, send data to PE0.
2034	IF ( ind(1) /= -9999 ) THEN
2035	CALL MPI_SEND( local_2d(nys,nzb_do), ngp, &
2036	MPI_REAL, 0, 1, comm2d, ierr )
2037	ENDIF
2038	ENDIF
2039	!
2040	!-- A barrier has to be set, because otherwise some PEs may
2041	!-- proceed too fast so that PE0 may receive wrong data on
2042	!-- tag 0
2043	CALL MPI_BARRIER( comm2d, ierr )
2044	ENDIF
2045
2046	ENDIF
2047	#else
2048	#if defined( __netcdf )
2049	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2050	id_var_do2d(av,ivar), &
2051	local_2d(nys:nyn,nzb_do:nzt_do), &
2052	start = (/ is, 1, 1, do2d_xz_time_count(av) /), &
2053	count = (/ 1, ny+1, nzt_do-nzb_do+1, 1 /) )
2054	CALL netcdf_handle_error( 'data_output_2d', 452 )
2055	#endif
2056	#endif
2057
2058	END SELECT
2059
2060	is = is + 1
2061	ENDDO loop1
2062
2063	!
2064	!-- For parallel output, all data were collected before on a local array
2065	!-- and are written now to the netcdf file. This must be done to increase
2066	!-- the performance of the parallel output.
2067	#if defined( __netcdf )
2068	IF ( netcdf_data_format > 4 ) THEN
2069
2070	SELECT CASE ( mode )
2071
2072	CASE ( 'xy' )
2073	IF ( two_d ) THEN
2074	nis = 1
2075	two_d = .FALSE.
2076	ELSE
2077	nis = ns
2078	ENDIF
2079	!
2080	!-- Do not output redundant ghost point data except for the
2081	!-- boundaries of the total domain.
2082	! IF ( nxr == nx .AND. nyn /= ny ) THEN
2083	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2084	! id_var_do2d(av,ivar), &
2085	! local_2d_sections(nxl:nxr+1, &
2086	! nys:nyn,1:nis), &
2087	! start = (/ nxl+1, nys+1, 1, &
2088	! do2d_xy_time_count(av) /), &
2089	! count = (/ nxr-nxl+2, &
2090	! nyn-nys+1, nis, 1 &
2091	! /) )
2092	! ELSEIF ( nxr /= nx .AND. nyn == ny ) THEN
2093	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2094	! id_var_do2d(av,ivar), &
2095	! local_2d_sections(nxl:nxr, &
2096	! nys:nyn+1,1:nis), &
2097	! start = (/ nxl+1, nys+1, 1, &
2098	! do2d_xy_time_count(av) /), &
2099	! count = (/ nxr-nxl+1, &
2100	! nyn-nys+2, nis, 1 &
2101	! /) )
2102	! ELSEIF ( nxr == nx .AND. nyn == ny ) THEN
2103	! nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2104	! id_var_do2d(av,ivar), &
2105	! local_2d_sections(nxl:nxr+1, &
2106	! nys:nyn+1,1:nis), &
2107	! start = (/ nxl+1, nys+1, 1, &
2108	! do2d_xy_time_count(av) /), &
2109	! count = (/ nxr-nxl+2, &
2110	! nyn-nys+2, nis, 1 &
2111	! /) )
2112	! ELSE
2113	nc_stat = NF90_PUT_VAR( id_set_xy(av), &
2114	id_var_do2d(av,ivar), &
2115	local_2d_sections(nxl:nxr, &
2116	nys:nyn,1:nis), &
2117	start = (/ nxl+1, nys+1, 1, &
2118	do2d_xy_time_count(av) /), &
2119	count = (/ nxr-nxl+1, &
2120	nyn-nys+1, nis, 1 &
2121	/) )
2122	! ENDIF
2123
2124	CALL netcdf_handle_error( 'data_output_2d', 55 )
2125
2126	CASE ( 'xz' )
2127	!
2128	!-- First, all PEs get the information of all cross-sections.
2129	!-- Then the data are written to the output file by all PEs
2130	!-- while NF90_COLLECTIVE is set in subroutine
2131	!-- define_netcdf_header. Although redundant information are
2132	!-- written to the output file in that case, the performance
2133	!-- is significantly better compared to the case where only
2134	!-- the first row of PEs in x-direction (myidx = 0) is given
2135	!-- the output while NF90_INDEPENDENT is set.
2136	IF ( npey /= 1 ) THEN
2137
2138	#if defined( __parallel )
2139	!
2140	!-- Distribute data over all PEs along y
2141	ngp = ( nxr-nxl+1 ) * ( nzt_do-nzb_do+1 ) * ns
2142	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
2143	CALL MPI_ALLREDUCE( local_2d_sections_l(nxl,1,nzb_do), &
2144	local_2d_sections(nxl,1,nzb_do), &
2145	ngp, MPI_REAL, MPI_SUM, comm1dy, &
2146	ierr )
2147	#else
2148	local_2d_sections = local_2d_sections_l
2149	#endif
2150	ENDIF
2151	!
2152	!-- Do not output redundant ghost point data except for the
2153	!-- boundaries of the total domain.
2154	! IF ( nxr == nx ) THEN
2155	! nc_stat = NF90_PUT_VAR( id_set_xz(av), &
2156	! id_var_do2d(av,ivar), &
2157	! local_2d_sections(nxl:nxr+1,1:ns, &
2158	! nzb_do:nzt_do), &
2159	! start = (/ nxl+1, 1, 1, &
2160	! do2d_xz_time_count(av) /), &
2161	! count = (/ nxr-nxl+2, ns, nzt_do-nzb_do+1, &
2162	! 1 /) )
2163	! ELSE
2164	nc_stat = NF90_PUT_VAR( id_set_xz(av), &
2165	id_var_do2d(av,ivar), &
2166	local_2d_sections(nxl:nxr,1:ns, &
2167	nzb_do:nzt_do), &
2168	start = (/ nxl+1, 1, 1, &
2169	do2d_xz_time_count(av) /), &
2170	count = (/ nxr-nxl+1, ns, nzt_do-nzb_do+1, &
2171	1 /) )
2172	! ENDIF
2173
2174	CALL netcdf_handle_error( 'data_output_2d', 57 )
2175
2176	CASE ( 'yz' )
2177	!
2178	!-- First, all PEs get the information of all cross-sections.
2179	!-- Then the data are written to the output file by all PEs
2180	!-- while NF90_COLLECTIVE is set in subroutine
2181	!-- define_netcdf_header. Although redundant information are
2182	!-- written to the output file in that case, the performance
2183	!-- is significantly better compared to the case where only
2184	!-- the first row of PEs in y-direction (myidy = 0) is given
2185	!-- the output while NF90_INDEPENDENT is set.
2186	IF ( npex /= 1 ) THEN
2187
2188	#if defined( __parallel )
2189	!
2190	!-- Distribute data over all PEs along x
2191	ngp = ( nyn-nys+1 ) * ( nzt-nzb + 2 ) * ns
2192	IF ( collective_wait ) CALL MPI_BARRIER( comm2d, ierr )
2193	CALL MPI_ALLREDUCE( local_2d_sections_l(1,nys,nzb_do), &
2194	local_2d_sections(1,nys,nzb_do), &
2195	ngp, MPI_REAL, MPI_SUM, comm1dx, &
2196	ierr )
2197	#else
2198	local_2d_sections = local_2d_sections_l
2199	#endif
2200	ENDIF
2201	!
2202	!-- Do not output redundant ghost point data except for the
2203	!-- boundaries of the total domain.
2204	! IF ( nyn == ny ) THEN
2205	! nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2206	! id_var_do2d(av,ivar), &
2207	! local_2d_sections(1:ns, &
2208	! nys:nyn+1,nzb_do:nzt_do), &
2209	! start = (/ 1, nys+1, 1, &
2210	! do2d_yz_time_count(av) /), &
2211	! count = (/ ns, nyn-nys+2, &
2212	! nzt_do-nzb_do+1, 1 /) )
2213	! ELSE
2214	nc_stat = NF90_PUT_VAR( id_set_yz(av), &
2215	id_var_do2d(av,ivar), &
2216	local_2d_sections(1:ns,nys:nyn, &
2217	nzb_do:nzt_do), &
2218	start = (/ 1, nys+1, 1, &
2219	do2d_yz_time_count(av) /), &
2220	count = (/ ns, nyn-nys+1, &
2221	nzt_do-nzb_do+1, 1 /) )
2222	! ENDIF
2223
2224	CALL netcdf_handle_error( 'data_output_2d', 60 )
2225
2226	CASE DEFAULT
2227	message_string = 'unknown cross-section: ' // TRIM( mode )
2228	CALL message( 'data_output_2d', 'PA0180', 1, 2, 0, 6, 0 )
2229
2230	END SELECT
2231
2232	ENDIF
2233	#endif
2234	ENDIF
2235
2236	ivar = ivar + 1
2237	l = MAX( 2, LEN_TRIM( do2d(av,ivar) ) )
2238	do2d_mode = do2d(av,ivar)(l-1:l)
2239
2240	ENDDO
2241
2242	!
2243	!-- Deallocate temporary arrays.
2244	IF ( ALLOCATED( level_z ) ) DEALLOCATE( level_z )
2245	IF ( netcdf_data_format > 4 ) THEN
2246	DEALLOCATE( local_pf, local_2d, local_2d_sections )
2247	IF( mode == 'xz' .OR. mode == 'yz' ) DEALLOCATE( local_2d_sections_l )
2248	ENDIF
2249	#if defined( __parallel )
2250	IF ( .NOT. data_output_2d_on_each_pe .AND. myid == 0 ) THEN
2251	DEALLOCATE( total_2d )
2252	ENDIF
2253	#endif
2254
2255	!
2256	!-- Close plot output file.
2257	file_id = 20 + s_ind
2258
2259	IF ( data_output_2d_on_each_pe ) THEN
2260	DO i = 0, io_blocks-1
2261	IF ( i == io_group ) THEN
2262	CALL close_file( file_id )
2263	ENDIF
2264	#if defined( __parallel )
2265	CALL MPI_BARRIER( comm2d, ierr )
2266	#endif
2267	ENDDO
2268	ELSE
2269	IF ( myid == 0 ) CALL close_file( file_id )
2270	ENDIF
2271
2272	CALL cpu_log( log_point(3), 'data_output_2d', 'stop' )
2273
2274	END SUBROUTINE data_output_2d

Note: See TracBrowser for help on using the repository browser.

Download in other formats:

| Impressum | ©Leibniz Universität Hannover |